PureBytes Links
Trading Reference Links
|
Tom,
Thursday, August 21, 2008, 11:30:43 PM, you wrote:
psb> Yuki,
psb> (This is just as true of rule-based systems as it is of discretionary
psb> trading; there is no difference at all in this respect between the
psb> two types of trading. None. Zero. The system without drawdowns has
psb> yet to be designed. And system traders don't also need discipline,
psb> and just as much, if not more? C'mon.)
psb> I wasn't saying that rule based didn't have drawdowns, I was
psb> saying that for some people it is easier to follow rules when
psb> things are going bad than to rely on their own judgment or feelings.
I don't think it's any easier at all. Rules-based systematic traders
are not stupid. They know they don't have any mathematical edge,
only an edge that for some period of time in the past, both in sample
and OOS, has tended to persist. But if they are at all smart, and I
suspect they are, they know this doesn't mean jack &^%$ as far as the
future goes. And knowing that, they take a HUGE deep breath when the
system suffers a major drawdown, and they ask themselves whether this
is the end, or just a speed bump. And reckoning that it is a speed
bump and saying damn the torpedoes doesn't *make* it a speed bump.
And because they know that, they doubt, they overrule, they do all
sorts of human-type things. It is *not* easy to stay the course when
you are taking hits and you have no sure mathematical edge. It is
anything but easy.
Now, I will grant you that if you trade a system based on human
psychology more than on optimizing or data mining, it might be easier
to keep the faith, because human behavior -- in general -- does not
change; it just fluctuates.
psb> (I can't believe I'm reading this. You are going to sit there and
psb> type to the world that 'thoroughly tested ... rules' are equivalent
psb> to a mathematical edge? You have to be joking, right? Mathematical
psb> edges don't go on tilt and *stay* on tilt forever. Trading systems
psb> (which are NOT mathematical edges, but assumptions about what might
psb> prove to be edges -- and almost surely edges with expiration dates --
psb> based on historical research) do just that)
psb> I don't believe that system trading is the same as a
psb> mathematical edge, just that a rules based program that has been
psb> tested completely gives you a foundation which is like relying on a mathematical edge.
This would be the Long Term Capital Management foundation, then? ^_^
Seriously, what kind of reliance can you put on something that, the
best system designers all admit, is bound to break down over time?
Some, for sure. But when the going gets rough? If one doesn't doubt
at those junctures, then one is actually not too bright.
psb> (As an aside, the only game in the casino that I know of that can be
psb> tilted in favor of the player is Blackjack, and I sure wouldn't want
psb> to try and make a living at it for quite a few reasons that should be
psb> obvious to the casual observer.)
psb>
psb> You must not be aware that there are video poker machines that
psb> have a positive mathematical edge some as high as 2%. Of course,
psb> the 2% machines are almost all gone. Look for books written by
psb> Bob Dancer.There are so many people making a living playing
psb> video poker that the casinos have been blackballing some video
psb> poker players. They also change the payouts on the video poker
psb> machines to below 100% return or increase the dollar per point on the cash back.
Most of the machines will be set to return some fraction of 100
percent, even if some are set to pay more. Unless you know which are
which, I doubt if you can figure it out. And I really, seriously,
doubt that "many people" (a slippery concept if there ever was one
for sure) are making a living playing video poker. Rest assured,
such an edge, if it exists, will be obliterated without prejudice by
the folks who own the machines. Life expectancy in that "business"
can probably be measured in "insect years". (I'm watching the
billions [trillions?] of cicadas drop from the skies around here, as
they do every late August and September, after putting up a racket
that will make your ears ring.)
psb> (When you lead with "the problem with discretionary trading is ... "
psb> it negates your disclaimer in the paragraph above. Certainly you are
psb> saying that rules are superior, because you claim that discretionary
psb> traders will have bad 'cycles' (as if rule-based traders will not,
psb> certainly what you are implying). And you are entitled to that
psb> opinion. You're just not entitled to say that it isn't your opinion
psb> when you very clearly state what you have stated.)
psb> You are correct. I shouldn't have lead with that statement. I
psb> should have said that "the problem for me was that with discretionary trading........"
psb> Thanks for your reply,
Well, you got me to post, something I do in California Cypress years
recently. ^_^
To put this to bed (for me), I would agree with you if you are
considering naive traders that either got a system from someone else,
or lucked onto their own system, not really knowing what they were
doing in the design phase. But anyone who really knows the score is
going to find systematic trading just as hard, if not *harder* than
"fly by the seat of your pants" trading. And when you are flying by
the seat of your pants, and the aircraft catches on fire, you are
basically free to jump out and pull the ripcord (kill the trade) when
you feel you should. When you are bound to rules, you have to ride
that airplane into the ground, don't you? And maybe if you survive,
you have to be willing to climb into really suspect airplanes for
another go. Doesn't sound too easy to me, but then I'm the cautious
type.
Yuki (licensed pilot -- small aircraft)
P.S. Hey Yanks, we rule in women's softball! ^_-
------------------------------------
Please note that this group is for discussion between users only.
To get support from AmiBroker please send an e-mail directly to
SUPPORT {at} amibroker.com
For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
http://www.amibroker.com/devlog/
For other support material please check also:
http://www.amibroker.com/support.html
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/amibroker/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/amibroker/join
(Yahoo! ID required)
<*> To change settings via email:
mailto:amibroker-digest@xxxxxxxxxxxxxxx
mailto:amibroker-fullfeatured@xxxxxxxxxxxxxxx
<*> To unsubscribe from this group, send an email to:
amibroker-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|